On Jul 20, 2009, at 9:09 AM, Dave Love wrote:
> > you should compile openmpi with each pf intel and gfortran seperatly
> > and install each of them in a separate location, and use mpi-
> selector
> > to select one.
>
> What, precisely, requires that, at least if you can recompile the MPI
> program with appropriate options? (Presumably it's features of the
> Fortran/C interfacing and/or Fortran runtime, but the former may be
> influenced by compilation options, and I'd hope the glue didn't
> require
> the compiler runtime -- the Intel compiler is on the list to check.)
>
See https://svn.open-mpi.org/source/xref/ompi_1.3/README#257.
> It's obviously of interest to those of us facing combinatorial
> explosion
> of libraries we're expected to install.
>
Indeed. In OMPI, we tried to make this as simple as possible. But
unless you use specific compiler options to hide their differences, it
isn't possible and is beyond our purview to fix. :-( (similar
situation with the C++ bindings)
> Also, is there any reason to use mpi-selector rather than switcher?
>
Nope -- they do about the same thing.
--
Jeff Squyres
jsquyres_at_[hidden]
|
![[Flag]](../../../../../community/mirrors/flag-us.gif){kind=small}